Systems, methods and articles of manufacture for minimizing parallax based on depth map generation are described herein. Embodiments of the present disclosure relate to cameras positioned on a camera assembly that capture digital photographic images. The digital photographic images are stitched together into a panoramic image. In order to minimize the parallax while stitching, the procedure includes estimating a distance of a pixel corresponding to a scene point on the scene surface from a panoramic center of projection positioned on the camera assembly to the scene surface. A depth map is generated that provides the estimated depth of the pixel. Based on the estimated depth, the pixel may be projected onto each camera. A panoramic pixel corresponding to the scene point may be obtained from the values of each projected pixel. The panoramic pixel may then be mapped onto the panoramic image with parallax minimized.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method for estimating a depth of each pixel corresponding to a scene point on a scene surface, comprising: identifying, by a processing device, a plurality of scan lines corresponding to a position of a corresponding ranging device positioned on a camera assembly as the camera assembly is travelling along a travel path, wherein each scan line has a starting point at the position of the corresponding ranging device and an end point at a position of the corresponding scene point on the scene surface; generating, by the processing device, a spherical coordinate system that is centered at a panoramic center of projection for the panoramic image, wherein the spherical coordinate system is configured to extend beyond the scene surface based on the plurality of scan lines; partitioning, by the processing device, the spherical coordinate system into a plurality of volumetric pixels; assigning, by the processing device, a depth value from a plurality of depth values to each pixel that is associated with each volumetric pixel; and determining, by the processing device, a location of the scene surface corresponding to the travel path of the camera assembly based on the depth value assigned to each pixel corresponding to each scene point on the scene surface.
2. The method of claim 1 , further comprising: generating a depth map based on the scan line distance for each scan line, wherein the depth map is configured to provide distance data along the scene surface relative to the travel path of the camera assembly.
3. The method of claim 1 , further comprising: surveying each volumetric pixel to determine whether each volumetric pixel is located in an internal region or an external region, wherein the internal region is located internal to the scene surface and the external region is located external to the scene surface.
4. The method of claim 3 , further comprising: identifying a first outside volumetric pixel located along a line of volumetric pixels radiating outward towards the scene surface from the position of the corresponding ranging device that is a first volumetric pixel along the line of volumetric pixels that is located outside the scene surface.
5. The method of claim 1 , wherein the assigning comprises: assigning a depth value to each pixel that corresponds to each volumetric pixel that is from an identified finite quantity of depth values.
6. The method of claim 1 , wherein the identifying comprises: receiving the plurality of scan lines from a plurality of lasers, wherein each laser is positioned on the camera assembly, wherein the position of each laser with respect to each camera positioned on the camera assembly is a known value.
7. A method for minimizing parallax from a panoramic image created by stitching together a plurality of digital photographic images captured from a plurality of cameras positioned on a camera assembly, comprising: accessing, by a processing device, the plurality of digital photographic images taken of a scene surface, wherein each digital photographic image is configured to be merged with one or more other images of the plurality of digital photographic images to form the panoramic image; estimating, by the processing device, a depth of each pixel corresponding to a scene point on the scene surface from a panoramic center of projection positioned on the camera assembly to the scene surface as the camera assembly is travelling along a travel path; projecting, by the processing device, the pixel corresponding to the scene point onto each camera of the plurality of cameras based on the estimated depth of each pixel; obtaining, by the processing device, a panoramic pixel corresponding to the scene point from each camera from the plurality of cameras based on a value of each projected pixel that corresponds to the scene point; and mapping, by the processing device, the panoramic pixel obtained from each camera onto the panoramic image.
8. The method of claim 7 , further comprising: stitching the plurality of digital photographic images together to generate the panoramic image of the scene surface.
9. The method of claim 7 , wherein the estimating comprises: deriving the depth of each pixel corresponding to the scene point on the scene surface from a depth map.
10. The method of claim 7 , wherein the obtaining comprises: combining each value of each projected pixel that corresponds to the scene point to obtain the panoramic pixel corresponding to the scene point.
11. A system for estimating a depth of each pixel corresponding to a scene point on a scene surface, comprising: a scan line identifier that is configured to provide a plurality of scan lines corresponding to a position of a corresponding ranging device positioned on a camera assembly as the camera assembly is travelling along a travel path, wherein each scan line has a starting point at the position of the corresponding ranging device and an end point at a position of the corresponding scene point on the scene surface; a spherical coordinate system generator that is configured to generate a spherical coordinate system that is centered at a panoramic center of projection for the panoramic image, wherein the spherical coordinate system is configured to extend beyond the scene surface based on the plurality of scan lines; a partitioning module that is configured to partition the spherical coordinate system into a plurality of volumetric pixels; an assignment module that is configured to assign a depth value from a plurality of depth values to each pixel that is associated with each volumetric pixel; and a determining module that is configured to determine a location of the scene surface corresponding to the travel path of the camera assembly based on the depth value assigned to each pixel corresponding to each scene point on the scene surface, wherein each of the scan line identifier, the spherical coordinate system generator, the partitioning module, the assignment module, and the determining module are implemented on a processing device.
12. The system of claim 11 , further comprising: a depth map generator that is configured to generate a depth map based on the scan line distance for each scan line, wherein the depth map is configured to provide distance data along the scene surface relative to the travel path of the camera assembly.
13. The system of claim 11 , further comprising: a surveying module that is configured to survey each volumetric pixel to determine whether each volumetric pixel is located in an internal region or an external region, wherein the internal region is located internal to the scene surface and the external region is located external to the scene surface.
14. The system of claim 13 , further comprising: an identification module that is configured to identify a first outside volumetric pixel located along a line of volumetric pixels radiating outward towards the scene surface from the position of the corresponding ranging device that is a first volumetric pixel along the line of volumetric pixels that is located outside the scene surface.
15. The system of claim 11 , wherein the assignment module is further configured to assign a depth value to each pixel that corresponds to each volumetric pixel from an identified quantity of finite depth values.
16. The system of claim 11 , wherein the identifying module comprises: a receiving module that is configured to receive the plurality of scan lines from a plurality of lasers, wherein each laser is positioned on the camera assembly, wherein the position of each laser with respect to each camera positioned on the camera assembly is a known value.
17. A system for minimizing parallax from a panoramic image created by stitching together a plurality of digital photographic images captured from a plurality of cameras positioned on a camera assembly, comprising: an accessing module that is configured to access the plurality of digital photographic images taken of a scene surface, wherein each digital photographic image is configured to be merged with one or more other images of the plurality of digital photographic images to form the panoramic image; an estimating module that is configured to estimate a depth of each pixel corresponding to a scene point on the scene surface from a panoramic center of projection positioned on the camera assembly to the scene surface as the camera assembly is travelling along a travel path; a projecting module that is configured to project the pixel corresponding to the scene point onto each camera of the plurality of cameras based on the estimated depth of each pixel; an obtaining module that is configured to obtain a panoramic pixel corresponding to the scene point from each camera from the plurality of cameras based on a value of each projected pixel that corresponds to the scene point; and a mapping module that is configured to map the panoramic pixel obtained from each camera onto the panoramic image, wherein the accessing module, the estimating module, the projecting module, the obtaining module, and the mapping module are implemented on a processing device.
18. The system of claim 17 , further comprising: a stitching module that is configured to stitch the plurality of digital photographic images together to generate the panoramic image of the scene surface.
19. The system of claim 17 , wherein the estimating module is further configured to derive the depth of each pixel corresponding to the scene point on the scene surface from a depth map.
20. The system of claim 17 , wherein the obtainment module is further configured to combine each value of each projected pixel that corresponds to the scene point to obtain the panoramic pixel corresponding to the scene point.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
November 14, 2012
April 21, 2015
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.